AI-200 Data Management Services And Vector Search Practice Test 7 — 15 Questions

Question 1

You are designing an enterprise RAG application that stores vector embeddings in Azure AI Search. You want to ensure that if the primary Azure region goes down, your search service automatically fails over with minimal data loss. How should you design your geo-replication strategy?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Configure geo-replication with replica services in a paired Azure region, enabling read-only failover and re-indexing capabilities. Deploying Azure AI Search with geo-replication across paired Azure regions ensures read availability, while writes can be resubmitted to the primary region upon recovery.

Answer

Export search indexes to CSV files hourly and upload them manually

Answer

Rely on Azure Blob Storage geo-redundancy to automatically mirror search index memory

Answer

Configure synchronous multi-region write master locks across 10 regions simultaneously

Question 2

When configuring vector search in Azure Cosmos DB for NoSQL, which TWO steps are required to prepare a container for vector search before inserting documents? (Choose two)

Accepted Answer

Define the vectorEmbeddingPolicy at container creation or update. Preparing a Cosmos DB container for vector search requires adding a vector embedding policy and defining vector indexes in the indexing policy.

Answer

Mount an NFS network share for vector file storage

Answer

Install the pgvector extension on the container

Answer

Create a relational foreign key constraint to Azure SQL

Question 3

You are configuring an Azure AI Search vector index for a retrieval-augmented generation (RAG) application. You need to select an algorithm for vector similarity search that balances recall speed with accuracy while allowing approximate nearest neighbor (ANN) search. Which vector search algorithm configuration should you select?

Accepted Answer

Hierarchical Navigable Small World (hnsw). HNSW (Hierarchical Navigable Small World) is the industry-standard algorithm used by Azure AI Search for approximate nearest neighbor (ANN) vector search, providing an optimal balance between fast retrieval speed and high recall.

Answer

Exhaustive K-Nearest Neighbor (exhaustiveKnn)

Answer

Locality-Sensitive Hashing (lsh)

Answer

Inverted File with Product Quantization (ivf-pq)

Question 4

You are optimizing an enterprise search architecture using Azure AI Search. Which THREE mechanisms help control latency and resource consumption during heavy vector query loads? (Choose three)

Accepted Answer

Tuning HNSW efSearch parameter to balance recall speed and latency. To control latency and resource consumption during heavy vector loads, you can scale out replicas, apply scalar quantization, and tune HNSW efSearch parameters.

Answer

Disabling caching and increasing index document count per partition beyond limits

Answer

Running exhaustive k-NN searches for every query by default

Question 5

You are provisioning an Azure AI Search service to support hybrid search, combining full-text lexical search and vector search. Which tier is the minimum required to support vector search capabilities at enterprise scale?

Accepted Answer

Basic tier. Azure AI Search supports vector search starting from the Basic tier, though Standard tiers (S1, S2, etc.) and Storage Optimized tiers are recommended for production enterprise scale.

Answer

Free tier

Answer

Storage Optimized L1

Answer

Standard 3 High Density (S3 HD)

Question 6

You are implementing a hybrid search solution in Azure AI Search that combines BM25 keyword search scores and HNSW vector search scores. By default, how does Azure AI Search normalize and combine these disparate score spaces before applying the RRF algorithm?

Accepted Answer

Reciprocal Rank Fusion (RRF) ranking combination. Azure AI Search uses Reciprocal Rank Fusion (RRF) to combine scores from different query types (such as text and vector) by normalizing ranks rather than raw scores, mitigating the scale discrepancy between BM25 and cosine distance.

Answer

Min-max feature scaling on raw scores

Answer

Softmax probability conversion

Answer

Z-score standardization on vector distances

Question 7

You are storing embeddings generated by text-embedding-3-small in Azure Cosmos DB for NoSQL. Which indexing policy property must be explicitly configured to enable vector search queries on your container?

Accepted Answer

Vector embedding policy and vector indexes. To run vector search queries in Azure Cosmos DB for NoSQL, the container's indexing policy must include a dedicated vector embedding path with a datatype of 'vector'.

Answer

Full-text search indexing policy

Answer

Composite indexes containing string fields

Answer

Included paths with geospatial indexes

Question 8

You are designing a multi-tenant enterprise RAG system using Azure AI Search. Each tenant requires isolated document retrieval. Which feature should you implement to enforce data security while leveraging vector search?

Accepted Answer

Security trimming using OData filter expressions with vector queries. Security trimming in Azure AI Search allows filtering search results based on user permissions by applying document-level access control lists (ACLs) using OData filters during the search query execution.

Answer

Role-Based Access Control (RBAC) at the index level

Answer

Dedicated search services per tenant without filters

Answer

Semantic ranker query throttling

Question 9

When storing vector embeddings in Azure Cosmos DB for MongoDB (vCore), which distance metric is natively supported when creating a vector search index?

Accepted Answer

Cosine distance (cos). Azure Cosmos DB for MongoDB (vCore) supports cosine distance, inner product, and Euclidean distance for vector search indexing.

Answer

Levenshtein distance

Answer

Jaccard distance

Answer

Hamming distance

Question 10

You are integrating Azure OpenAI Service's text-embedding-ada-002 model with Azure AI Search. You want to automate the generation of embeddings directly inside the search service during data ingestion. Which component of Azure AI Search should you configure?

Accepted Answer

Azure AI Search Indexer with integrated vectorization. Indexers in Azure AI Search support integrated vectorization, allowing you to connect to Azure OpenAI to automatically chunk text and generate embeddings during the indexing pipeline.

Answer

Semantic ranker configuration profile

Answer

Azure Function scheduled trigger for batch updates

Answer

Custom Web API skill in Azure AI Services enrichment pipeline

Question 11

You are configuring vector search on an Azure AI Search index using a custom vectorizer that points to an external embedding endpoint. Which authentication method is recommended and natively supported for securely connecting Azure AI Search to the Azure OpenAI endpoint?

Accepted Answer

Managed Identity (system-assigned or user-assigned). Azure AI Search supports managed identities (both system-assigned and user-assigned) to securely authenticate against Azure OpenAI endpoints without storing API keys.

Answer

OAuth2 client credentials grant with Azure AD tenant federation

Answer

Shared Access Signature (SAS) tokens

Answer

Basic authentication with API keys stored in index metadata

Question 12

You are designing an application that stores vector embeddings in Azure SQL Database. Which data type should you use to store vector data natively in modern Azure SQL?

Accepted Answer

VECTOR native data type. Azure SQL Database supports the vector data type (along with native vector functions like vector_distance) to store and query high-dimensional embeddings efficiently.

Answer

VARCHAR(MAX) JSON string

Answer

FLOAT array collection

Answer

VARBINARY(MAX) blob

Question 13

You are configuring an Azure AI Search index with vector fields. You need to ensure that filtering operations (such as matching a category ID) are executed with 100% precision before or alongside the vector search. Which vector filter mode should you choose?

Accepted Answer

preFilter mode. Pre-filtering (preFilter) applies the OData filter before the vector search execution, ensuring that only documents matching the filter are considered in the vector similarity calculation.

Answer

approximateFilter mode

Answer

hybridFilter mode

Answer

postFilter mode

Question 14

You are troubleshooting a high memory consumption issue on an Azure AI Search service hosting large vector indexes. Which index configuration choice contributes most directly to high memory pressure when using the HNSW algorithm?

Accepted Answer

Storing raw uncompressed vectors in memory for HNSW graph traversal. HNSW indexes must load the entire vector graph structure into RAM to execute fast approximate nearest neighbor searches, making memory sizing critical.

Answer

Configuring high-frequency indexers

Answer

Enabling semantic ranking on text fields

Answer

Using exhaustiveKnn instead of HNSW

Question 15

You are querying an Azure Cosmos DB for NoSQL container using vector search. You execute a query using the VectorDistance() system function. Which query structure is required to properly execute this vector search?

Accepted Answer

SELECT TOP 10 c.id, VectorDistance(c.vectorProp, [0.1, 0.2]) AS score FROM c ORDER BY VectorDistance(c.vectorProp, [0.1, 0.2]). Vector search queries in Azure Cosmos DB for NoSQL require using the VectorDistance system function in the ORDER BY clause to sort by similarity and selecting the top results using TOP.

Answer

SELECT TOP 10 c.id FROM c WHERE VectorDistance(c.vectorProp, [0.1, 0.2]) > 0.9

Answer

JOIN vectorIndex ON c.id = vectorIndex.id

Answer

GROUP BY VectorDistance(c.vectorProp, [0.1, 0.2])